Author + information
- Received August 1, 2015
- Revision received August 21, 2015
- Accepted August 25, 2015
- Published online November 10, 2015.
- Jae-Sik Jang, MD, PhD∗,†,‡,
- John A. Spertus, MD, MPH∗,†∗ (, )
- Suzanne V. Arnold, MD, MHA∗,†,
- Ali Shafiq, MD∗,†,
- Anna Grodzinsky, MD∗,†,
- Timothy J. Fendler, MD∗,†,
- Adam C. Salisbury, MD, MSc∗,†,
- Fengming Tang, MS∗,
- Edward J. McNulty, MD§,
- J. Aaron Grantham, MD∗,†,
- David J. Cohen, MD, MSc∗,† and
- Amit P. Amin, MD, MSc‖,¶
- ∗Cardiovascular Outcomes Research, Saint Luke’s Mid America Heart Institute, Kansas City, Missouri
- †University of Missouri−Kansas City, Kansas City, Missouri
- ‡Department of Cardiology, Busan Paik Hospital, Inje University College of Medicine, Busan, Republic of Korea
- §Division of Cardiology, Kaiser Permanente Medical Center, San Francisco, California
- ‖Cardiovascular Division, Washington University School of Medicine, St. Louis, Missouri
- ¶Barnes-Jewish Hospital, St. Louis, Missouri
- ↵∗Reprint requests and correspondence:
Dr. John A. Spertus, Saint Luke’s Mid America Heart Institute, 4401 Wornall Road, Kansas City, Missouri 64111.
Background Up to 65% of patients with ST-segment elevation myocardial infarction (STEMI) have multivessel coronary artery disease (MVCAD). Long-term health status of STEMI patients after multivessel revascularization is unknown.
Objectives This study investigated the relationship between multivessel revascularization and health status outcomes (symptoms and quality of life [QoL]) in STEMI patients with MVCAD.
Methods Using a U.S. myocardial infarction registry and the Seattle Angina Questionnaire (SAQ), we determined the health status of patients with STEMI and MVCAD at the time of STEMI and 1 year later. We assessed the association of multivessel revascularization during index hospitalization with 1-year health status using multivariable linear regression analysis, and also examined demographic, clinical, and angiographic factors associated with multivessel revascularization.
Results Among 664 STEMI patients with MVCAD, 251 (38%) underwent multivessel revascularization. Most revascularizations were staged during the index hospitalization (64.1%), and 8.0% were staged after discharge, with 27.9% performed during primary percutaneous coronary intervention. Multivessel revascularization was associated with age and more diseased vessels. At 1 year, multivessel revascularization was independently associated with improved symptoms (4.5 points higher SAQ angina frequency score; 95% confidence interval [CI]: 1.0 to 7.9) and QoL (6.6 points higher SAQ QoL score; 95% CI: 2.7 to 10.6). One-year mortality was not different between those who did and did not undergo multivessel revascularization (3.6% vs. 3.4%; log-rank test p = 0.88).
Conclusions Multivessel revascularization improved angina and QoL in STEMI patients with MVCAD. Patient-centered outcomes should be considered in future trials of multivessel revascularization.
Among ST-segment elevation myocardial infarction (STEMI) patients, 40% to 65% have 1 or more significant coronary stenoses in addition to their culprit vessels (1). When and how to treat these nonculprit vessels is controversial (2). Options include revascularization at the time of primary percutaneous coronary intervention (PCI), later during the acute myocardial infarction (AMI) hospitalization, or medical therapy with or without subsequent elective revascularization after discharge. Current guidelines recommend that only the culprit vessel be treated during index PCI in the absence of hemodynamic instability after opening the culprit artery (3,4), with staged procedures being considered in case of symptoms or ischemia within days to weeks after primary PCI (5).
Recently, several observational studies have been published with conflicting results about the benefits of nonculprit PCI (6–8). These observational studies have been supplemented with randomized trials, including PRAMI (Preventive Angioplasty in Acute Myocardial Infarction) (9) and CvLPRIT (Complete versus Lesion only Primary-PCI Trial) (10), that have reported better composite outcomes in STEMI patients treated with complete revascularization at the time of primary PCI or during the index hospitalization. Importantly, these studies did not examine patient-reported outcomes, and there have been no data reported on the long-term health status outcomes of culprit-only or multivessel revascularization. This is particularly important as reduced angina and improved quality of life (QoL) could be important potential benefits of treating nonculprit flow-limiting lesions, similar to the results of studies in population of patients with stable coronary artery disease (11,12). To define the potential patient-centered benefits of complete revascularization, we sought to describe the patterns of treating noninfarct vessels, patient characteristics associated with multivessel revascularization, variation in practice across hospitals, and the independent association of multivessel revascularization with 1-year health-related QoL and mortality.
Study protocol and population
The TRIUMPH (Translational Research Investigating Underlying Disparities in Acute Myocardial Infarction Patients' Health Status) study is a prospective, multicenter cohort study of 4,340 AMI patients enrolled at 24 U.S. centers between April 2005 and December 2008. Patients were eligible for inclusion if they were ≥18 years of age and had an AMI, and elevated biomarkers with either electrocardiographic changes or symptoms consistent with the diagnosis (13). TRIUMPH was approved by the institutional research board at each participating site and written informed consent was obtained from all participants.
From the overall TRIUMPH cohort, we first identified patients with STEMI and multivessel coronary artery disease (MVCAD), which was defined as having at least 2 of 3 epicardial vessels with a stenosis ≥70% or left main stenosis ≥50%. Given the goal of examining practice patterns and long-term outcomes in STEMI patients with MVCAD, we excluded those with non−ST-segment elevation myocardial infarction (NSTEMI), prior coronary artery bypass grafting (CABG), and those who died during hospitalization.
Definition of multivessel revascularization
Multivessel revascularization was defined as revascularization of all major coronary stenoses during the index hospitalization or within 6 weeks of discharge with either PCI or CABG, such that there was no residual stenosis ≥70% in any major epicardial coronary artery. This definition included simultaneous nonculprit PCI during the index primary PCI, staged revascularizations during the index admission, and elective procedures performed within 6 weeks of discharge. For patients undergoing CABG during the index hospitalization or within 6 weeks of discharge, we assumed that complete revascularization had been performed. Conversely, we defined culprit-only revascularization as PCI to the culprit vessel only during the index hospitalization or multivessel PCI with residual untreated or unsuccessfully treated noninfarct artery stenoses ≥70% after the index primary PCI.
Disease-specific health status was prospectively assessed using the Seattle Angina Questionnaire (SAQ) at the time of patients’ index hospitalizations and at 1-year follow-up. The SAQ is a 19-item patient-reported health status instrument with a recall period of 4 weeks that quantifies 5 clinically relevant domains of coronary disease, including angina frequency (SAQ AF) and quality of life (SAQ QoL). Scores in these domains range from 0 to 100, with higher scores indicating fewer symptoms and better QoL (14). The SAQ has been demonstrated to be valid, reliable, and sensitive to clinical changes (14–17) and is also associated with subsequent rehospitalization and health care costs (18–20). As a secondary outcome of interest, we assessed all-cause mortality for a period of 1 year after the index admission using a combination of phone follow-up and the Social Security death master index. We also examined the rates of severe angina (defined as having more than 3 episodes of angina per week as defined by a SAQ AF score of ≤40), myocardial infarction, and repeat revascularization procedures (PCI or CABG) during the year after patients’ initial AMI.
Baseline clinical and demographic characteristics were compared between those with multivessel versus culprit-only revascularization using the Student t test or Mann-Whitney U tests for continuous variables and chi-square or Fisher exact test for categorical variables, as appropriate. To identify factors associated with multivessel revascularization, we used a multivariable, hierarchical (adjusting for site as a random effect) modified Poisson regression model with robust variance estimation (21). Factors included in the model were selected a priori, on the basis of published data review and clinical judgment, and included age; sex; race; history of atrial fibrillation, diabetes mellitus and congestive heart failure; left ventricular systolic function; number of diseased vessels; non−left anterior descending (LAD) artery as the culprit artery; development of in-hospital heart failure after the primary PCI; baseline creatinine; baseline hemoglobin; short form-12 physical component scale score at baseline; presence of angina before admission; financial status; and avoidance of health care services due to cost. Variation in the practice of multivessel revascularization across the study sites was assessed using the median rate ratio (MRR) (22,23).
The association of multivessel revascularization with 1-year health status outcomes was evaluated using hierarchical (adjusting for site as a random effect) multivariable linear regression models for each health status outcome (SAQ AF and SAQ QoL). Covariates for adjustment were selected a priori, on the basis of published data review and clinical judgment, and included baseline health status scores, age, sex, race, history of smoking, number of diseased vessels, non-LAD as the culprit artery, in-hospital heart failure after the primary PCI, baseline creatinine, baseline hemoglobin, and self-reported avoidance of health care services due to cost. Because these models included baseline health status, the regression coefficient for the “multivessel” versus culprit-only artery revascularization variable represented the difference between these 2 groups in the improvement of the health status scores from baseline to 1 year, adjusted for all covariates.
The associations of multivessel revascularization with 1-year mortality, myocardial infarction, and repeat revascularization were assessed using Kaplan-Meier survival analysis and the log-rank test. Finally, we performed a sensitivity analysis excluding patients undergoing CABG to determine whether our results were comparable in the PCI-only patients. All statistical analyses were conducted using SAS version 9.3 (SAS Institute Inc., Cary, North Carolina).
Two hundred twenty-four (33.7%) of the 664 patients did not have a follow-up SAQ score at 1 year. Outcomes data were missing equally in the 2 groups. Baseline characteristics were compared between those with versus without missing data and we sought to minimize the effect of selection bias (due to loss-to-follow-up) by constructing a nonparsimonious, multivariable logistic regression model to determine the probability of having missing data. We then weighted each of the observed patients by the inverse probability of the likelihood to having missing data (24) so that we preferentially weighted the experience of those most like the patients who were missing follow-up assessments.
Of the 4,340 patients enrolled in TRIUMPH, 537 patients with a history of prior CABG were excluded. We also excluded 2,218 patients with single vessel coronary disease or who did not undergo diagnostic angiography. We further excluded 406 NSTEMI patients and 4 patients who died before discharge, making our final analytic cohort 664 patients with STEMI and MVCAD (Figure 1). Among these, 413 (62%) underwent culprit-only PCI and 251 (38%) underwent multivessel revascularization. Of those undergoing complete revascularization, most were staged during index hospitalization (n = 161, 64.1%), 20 (8.0%) were performed shortly after discharge, and 70 (27.9%) were treated during their primary PCI. The mode of multivessel revascularization was percutaneous in 88% of patients. The baseline characteristics of those who underwent multivessel versus culprit-only revascularization are displayed in Table 1. Patients undergoing multivessel revascularization were younger, had less history of prior PCI, had higher peak troponin and hemoglobin levels, received more drug-eluting stents, had a greater number of diseased vessels, and were more likely to have proximal LAD artery involvement of the nonculprit artery.
Factors associated with multivessel revascularization
The factors independently associated with multivessel revascularization were the number of diseased vessels (relative risk: 1.31 per additional vessel; 95% confidence interval [CI]: 1.17 to 1.46) and age (Figure 2). Age displayed a nonlinear, inverse J-shaped association with multivessel revascularization. Whereas older patients were less likely to receive multivessel revascularization, as compared with patients around age 50 years, the youngest patients also showed a trend towards less multivessel revascularization, which was not statistically significant (Online Figure 1).
The average rate of multivessel revascularization across the population was 38%; however, we observed marked variation across study sites, ranging from 0% to >70% (Figure 3). The MRR, a statistical measure of variation, was 1.30, suggesting an average 30% greater likelihood of undergoing multivessel revascularization for patients with identical characteristics who present at one randomly selected TRIUMPH site as compared with another.
Multivessel versus culprit-only revascularization and health status outcomes
Among the 664 eligible patients, 224 (33.7%) did not have a follow-up SAQ score at 1 year. These data were missing equally in the 2 groups and were addressed with inverse probability weighting to preferentially weight the experience of patients most like those missing follow-up data. There were improvements in unadjusted health status from baseline to 1 year in both groups (Table 2), although these improvements were greater in the multivessel revascularization group. On average, the unadjusted SAQ AF score in the multivessel group was 4.09 points higher than the culprit-only group (95% CI: 0.56 to 7.63) (Figure 4A). The proportion of patients reporting symptoms of severe angina (i.e., more than 3 angina episodes per week) at 1 year was not significantly different between the 2 groups (4.4% vs. 6.3%; p = 0.34) (Table 2). The unadjusted SAQ QoL score was also significantly greater in the multivessel group by 5.18 points (95% CI: 1.17 to 9.19). After multivariable regression with inverse probability weighting to adjust for baseline patient characteristics, there was a significant difference in the degree of improvement from baseline to 1 year between the groups of patients who underwent complete, as compared with culprit-only, revascularization. The SAQ AF score was 4.45 points (95% CI: 0.99 to 7.91) higher and the SAQ QoL score was 6.63 points (95% CI: 2.67 to 10.59) (Figure 4A) higher with multivessel versus culprit-only revascularization. The interaction of the timing of complete revascularization (at the time of primary PCI, later in the hospital stay or shortly after discharge) was not significant (p = 0.64 and p = 0.69 for SAQ AF and SAQ QoL, respectively).
Multivessel versus culprit-only revascularization and clinical outcomes
There was no difference in 1-year mortality between those who did and did not undergo multivessel revascularization (3.6% vs. 3.4%; log-rank test p = 0.88) (Table 2, Online Figure 2). A total of 32 patients (9.1%) in the culprit-only group and 17 patients (7.5%) in the multivessel revascularization group underwent further revascularization during 12 months (log-rank test p = 0.50).
When these analyses were repeated in the PCI-only cohort, patients who underwent multivessel PCI had significant improvements in health status as compared with culprit-only PCI. The multivariable-adjusted difference in SAQ AF scores was 5.34 points (95% CI: 1.77 to 8.92) and in SAQ QoL scores it was 7.78 points (95% CI: 3.79 to 11.77) (Figure 4B).
To evaluate the health status differences between multivessel and culprit-only revascularization in STEMI patients, we analyzed a multicenter study of contemporary revascularization practice in the United States. We found that multivessel revascularization was performed in 38% of STEMI patients with MVCAD and that there was modest variation across hospitals. Patient characteristics independently associated with multivessel revascularization included age and a higher number of diseased vessels. Finally, multivessel revascularization was associated with improvements in both angina and QoL, but not mortality or repeat revascularization procedures, 1 year after discharge (Central Illustration). To our knowledge, this is the first study to evaluate patients’ perception of their symptoms and QoL after complete versus culprit-only revascularization in patients with STEMI and MVCAD and these data extend recent insights from clinical trials of a potential clinical benefit from complete revascularization by also suggesting an improvement in patients’ 1-year health status.
Several recent studies of different revascularization strategies in STEMI patients and MVCAD have been performed (2,25,26), with 2 randomized clinical trials showing better clinical outcomes with complete revascularization (9,10). The PRAMI study compared multivessel (complete) PCI versus culprit-only PCI at time of primary PCI in 465 STEMI patients with MVCAD and found that multivessel PCI decreased the incidence of the primary composite endpoint of cardiac death, nonfatal myocardial infarction, or refractory angina (hazard ratio [HR]: 0.35; 95% CI: 0.21 to 0.58; p < 0.001) (9). They also reported significantly lower refractory angina (HR: 0.35; 95% CI: 0.18 to 0.69; p = 0.002) and repeat revascularization (HR: 0.30; 95% CI: 0.17 to 0.56; p < 0.001) in the multivessel revascularization group. Another recent randomized trial, CvLPRIT (10), compared a strategy of complete in-hospital revascularization with culprit-only PCI and ischemia-based subsequent revascularization in 296 STEMI patients with MVCAD at time of primary PCI. They found that complete revascularization, including staged in-hospital procedure before discharge, was associated with a lower incidence of death, myocardial infarction, heart failure, or ischemia-driven revascularization at 12 months (10). Our findings extend with these trials by finding less angina and better disease-specific QoL at 1 year, although we were not adequately powered to detect a difference in mortality or subsequent revascularization.
Our study also revealed interesting patterns in the practice of selecting patients with MVCAD for multivessel revascularization during a STEMI hospitalization. Among diverse care settings throughout the United States, we found that the very old and very young patients were less likely to undergo multivessel revascularization. Similarly, those with a higher number of diseased vessels were more likely to undergo complete revascularization during their hospitalization, rather than at the time of primary reperfusion. Several other factors (e.g., history of prior PCI, troponin level, distribution of culprit vessels) that were significantly different between patients receiving different revascularization strategies were not independently correlated with multivessel revascularization within 6 weeks of an AMI. Unmeasured confounding or selection bias may have influenced our findings because we did not query the clinicians about why complete or culprit-only revascularization was performed. Our findings thus emphasize the need for future randomized trials to compare these 2 revascularization strategies and to explicitly measure the effects of treatment strategy on QoL, an outcome of critical importance to patients.
However, the majority of multivessel revascularizations in our study seemed to be driven largely by clinical factors, such as worsening clinical status and anatomic considerations, rather than by presenting sites or physician factors. In fact, the MRR showed only modest variability with only an average 30% variation in the likelihood of receiving multivessel revascularization for a given patient presenting to 1 hospital in our cohort versus another. These variations in care may stem from confusion in the existing published data surrounding the optimal treatment strategy for STEMI patients with MVCAD. Some of this variation might be attributable to the difficulty in interpreting previous studies with different timing and definition of staged and complete revascularization while examining the outcomes of complete revascularization (9,10,25–33) and emphasizes the opportunity to improve the consistency and quality of care for STEMI patients with MVCAD. Studies excluding staged procedure from either complete or culprit-only groups reported better outcomes in the culprit-only arm (34,35). Conversely, studies considering staged PCI, including staged in-hospital PCI procedures as a complete revascularization strategy, reported better outcomes with complete revascularization (9,33,36). Congruent with these findings, our definition of staged procedure included all settings in which multivessel revascularization was accomplished either in the hospital or shortly thereafter. Importantly, we found no significant interaction between different strategies of performing multivessel revascularization on outcomes, suggesting similar outcomes with any of the 3 approaches to complete revascularization.
Current guidelines do not recommend revascularization of the noninfarct artery unless indicated by hemodynamic instability or persistent ischemia after PCI of the supposed culprit lesion (3,4) and the clinicians following these patients in the outpatient setting may not recommend intervention for mildly symptomatic STEMI patients. This is supported by our findings that there was not a significantly higher prevalence of severe angina in the culprit-only group. However, patients with multivessel revascularization had a statistically and clinically significant 7-point greater improvement in SAQ QoL as compared with those undergoing culprit-only PCI, even after adjusting for baseline health status and other factors, and a slightly smaller, 4.5-point mean improvement in SAQ AF scores that were larger than those observed in the COURAGE (Clinical Outcomes Utilizing Revascularization and Aggressive Drug Evaluation) trial (37). One may therefore wonder why patients who have residual angina in the year after AMI might not receive more aggressive antianginal therapy or undergo revascularization of their untreated stenosis. Although our study cannot provide an explanation for this, there is known to be variability in the quality of follow-up care that may preclude more aggressive treatment of patients’ angina and, given our findings of better angina control and disease-specific QoL, we believe that more complete revascularization at the time of a STEMI might be warranted (38). To definitely prove this, however, randomized trials that systematically measure patients’ health status are needed.
Our study should be interpreted in the context of several potential limitations. First, in light of the observational nature of our data, it is not possible to establish a causal relationship between multivessel revascularization and improved QoL at 1 year. Patients in our study were likely aware of whether their residual coronary stenoses were fixed or not and this may have influenced their perceptions of recurrent chest pain or perceived QoL. Although these data suggest that health-related QoL is improved with multivessel revascularization in STEMI patients with MVCAD, unmeasured confounding or selection bias cannot be excluded and may have influenced our findings. Second, we did not query the clinicians as to why complete or culprit-only revascularization was performed and there may have been prognostically important considerations that were not included in our risk-adjusted models. A randomized trial would be needed to remove this confounding, although it is possible that patients in whom the physicians believed complete revascularization would be needed would not enroll such patients.
A third consideration in the interpretation of our findings is that one-third of patients were missing SAQ data at 1 year that may have introduced a selection bias, although the rates at which these data were missing did not differ between those with culprit-only and multivessel revascularization. A fourth concern is that an angiographic core laboratory was not used to assess the severity of CAD and angiographic data were obtained by chart abstraction of the dictated reports. Any misclassifications, however, might be expected to bias our study results toward the null. In addition, our secondary analysis of mortality outcomes had limited power to detect any differences in survival. Given the importance of survival, adequately powered randomized clinical trials will be needed to better define the impact of complete revascularization on subsequent mortality. Finally, we excluded patients who did not survive to hospital discharge, and, as such, the observed benefits of multivessel revascularization cannot be extrapolated to extremely sick patients.
In a large multicenter AMI registry, we found that multivessel revascularization in the setting of STEMI was common, varied by both patient characteristics and the treating hospital, and was strongly associated with a clinically significant improvement in both angina and QoL 1 year after treatment. Future studies of the potential benefits and harms of multivessel revascularization in STEMI patients should include both symptoms and health-related QoL outcomes so that more complete insights into the benefits of multivessel, as compared with culprit-only, revascularization can be assessed.
COMPETENCY IN MEDICAL KNOWLEDGE: Multivessel revascularization at time of primary PCI or during index hospitalization in patients with STEMI is associated with better symptom status and quality of life 1 year later.
TRANSLATIONAL OUTLOOK: Additional studies are needed to guide selection of patients with acute STEMI for immediate versus staged multivessel PCI.
For supplemental figures, please see the online version of this article.
The TRIUMPH study was funded by a grant from the NHLBI (P50 HL 077113) and CV Outcomes (Kansas City, Missouri). Drs. Shafiq, Grodzinsky, and Fendler received support from the National Heart, Lung, and Blood Institute of the National Institutes of Health under Award Number T32HL110837. Dr. Amin is funded via a comparative effectiveness research KM1 career development award from the Clinical and Translational Science Award (CTSA) program of the National Center for Advancing Translational Sciences of the National Institutes of Health, Grant Numbers UL1TR000448, KL2TR000450, TL1TR000449; and the National Cancer Institute of the National Institutes of Health, Grant Number 1KM1CA156708-01; and is a consultant with Teruno, The Medicines Company, and AstraZeneca.
Dr. Jang was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIP) (No. R13-2007-023-00000-0). Dr. Spertus owns the copyright to the Seattle Angina Questionnaire, KCCQ, and PAQ; and has equity in Health Outcome Sciences. Dr. Grantham has received grants, honoraria, speaking fees, and travel reimbursement from Medtronic, Boston Scientific, Abbott Vascular, and ASAHI Intecc; is a shareholder of and has intellectual property with Insysiv, LLC.
All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
Timothy D. Henry, MD, served as Guest Editor for this paper.
- Abbreviations and Acronyms
- angina frequency
- acute myocardial infarction
- coronary artery bypass grafting
- left anterior descending
- median rate ratio
- multivessel coronary artery disease
- percutaneous coronary intervention
- quality of life
- Seattle Angina Questionnaire
- ST-segment elevation myocardial infarction
- Received August 1, 2015.
- Revision received August 21, 2015.
- Accepted August 25, 2015.
- American College of Cardiology Foundation
- Sorajja P.,
- Gersh B.J.,
- Cox D.A.,
- et al.
- O'Gara P.,
- Kushner F.,
- Ascheim D.,
- et al.
- Steg P.,
- James S.,
- Atar D.,
- et al.
- Windecker S.,
- Kolh P.,
- Alfonso F.,
- et al.
- Gershlick A.H.,
- Khan J.N.,
- Kelly D.J.,
- et al.
- Pursnani S.,
- Korley F.,
- Gopaul R.,
- et al.
- Thygesen K.,
- Alpert J.S.,
- White H.D.,
- et al.
- Spertus J.A.,
- Winder J.A.,
- Dewhurst T.A.,
- et al.
- Arnold S.V.,
- Morrow D.A.,
- Lei Y.,
- et al.
- Spertus J.A.,
- Jones P.,
- McDonell M.,
- et al.
- Zou G.
- Larsen K.,
- Merlo J.
- Politi L.,
- Sgura F.,
- Rossi R.,
- et al.
- Hannan E.L.,
- Samadashvili Z.,
- Walford G.,
- et al.
- Toma M.,
- Buller C.E.,
- Westerhout C.M.,
- et al.
- Cavender M.A.,
- Milford-Beland S.,
- Roe M.T.,
- et al.
- Bainey K.R.,
- Mehta S.R.,
- Lai T.,
- Welsh R.C.
- Spertus J.A.,
- Salisbury A.C.,
- Jones P.G.,
- et al.